JP2016519519A - Video signal processing method and apparatus - Google Patents

Abstract

A video signal processing method for improving the coding efficiency of depth data is provided. A video signal processing method according to the present invention acquires a depth prediction value of a current block, restores a depth residual for each sample of the current block according to an SDC mode indicator, and obtains a depth prediction value and a restored depth register. Restore depth value of current block using dual. Coding efficiency for the depth data can be improved by adaptively using the SDC mode according to the SDC mode indicator and further using the SDC mode and / or the depth look-up table. [Selection] Figure 3

Description

  The present invention relates to a video signal coding method and apparatus.

  Compression encoding refers to a series of signal processing techniques for transmitting digitized information via a communication line or storing the information in a form suitable for a storage medium. Audio, video, text, and the like exist as the targets of compression coding. In particular, a technique for performing compression coding for video is called video video compression. A general feature of multi-view video images is that they have spatial redundancy, temporal redundancy, and inter-viewpoint redundancy.

  An object of the present invention is to increase coding efficiency of video signals, particularly depth data.

  To achieve the above object, the present invention obtains a depth prediction value of a current block, restores a depth residual for each sample of the current block according to an SDC mode indicator, and stores the depth prediction value and the restored depth register. The depth value of the current block is restored using a dual.

  The SDC mode indicator according to the present invention refers to a flag indicating whether the current block is coded in the SDC mode, and the SDC mode is a depth residual for a plurality of samples in the current block. It means a method of coding in one depth residual.

  When the SDC mode indicator according to the present invention indicates that the current block is coded in the SDC mode, the depth residual of the current block is restored using residual coding information.

  The residual coding information according to the present invention includes depth absolute absolute values and depth residual code information.

  The depth residual according to the present invention means a difference between the average value of the depth values of the current block and the average value of the depth prediction values of the current block.

  The depth residual according to the present invention means an average value of the depth residual of the i-th sample obtained from the difference between the depth value of the i-th sample of the current block and the depth predicted value of the i-th sample. .

  If the SDC mode indicator according to the present invention indicates that the current block is coded in SDC mode, the depth residual is restored using a depth lookup table.

  The depth residual according to the present invention derives a difference index using the absolute value of the depth residual and the sign information of the depth residual, obtains the depth prediction average value of the current block, and obtains the depth prediction average value and the depth prediction value. A prediction index is obtained using a lookup table, a table depth value corresponding to an index obtained from the sum of the prediction index and the difference index is obtained from the depth lookup table, and the obtained table depth value is obtained. And a depth prediction average value.

  The prediction index according to the present invention is set as a table index assigned to the table depth value that minimizes the difference between the depth prediction average value and the table depth value in the depth lookup table.

  The effects of the video signal processing method and apparatus according to the present invention will be described as follows.

  According to at least one of the embodiments of the present invention, the coding efficiency of depth data can be improved by adaptively using the SDC mode using the SDC mode indicator.

  According to at least one of the embodiments of the present invention, in the SDC mode, it is possible to code one depth residual without coding the depth residual for all the samples in the current block, and vice versa. Since the quantization and inverse transformation processes are skipped, the depth residual coding efficiency can be improved.

  According to at least one of the embodiments of the present invention, it is possible to reduce errors due to round calculation by performing an average calculation after calculating a difference between the depth value of the current block and the depth prediction value of the current block. .

  According to at least one of the embodiments of the present invention, the number of bits required for coding the depth data can be reduced by converting the depth value into an index using the depth lookup table.

1 is a schematic block diagram of a video decoder 100 according to an embodiment to which the present invention is applied. 1 is an internal block diagram of a broadcast receiver to which a video decoder is applied according to an embodiment to which the present invention is applied. 6 is a flowchart illustrating a process of restoring a depth value of a current block according to an embodiment to which the present invention is applied. FIG. 5 is a diagram illustrating a method of encoding residual coding information when an embodiment to which the present invention is applied and no depth lookup table is used. FIG. 6 is a diagram illustrating a method for obtaining a depth residual of a current block using residual coding information when a depth look-up table is not used, according to an embodiment to which the present invention is applied. FIG. 5 is a diagram illustrating a method of encoding residual coding information when a depth lookup table is used according to an embodiment to which the present invention is applied. FIG. 5 is a diagram illustrating a method of restoring depth residual using residual coding information when a depth lookup table is used according to an embodiment to which the present invention is applied.

  In order to achieve the above object, a video signal processing method according to the present invention obtains a depth prediction value of a current block, restores a depth residual for each sample of the current block according to an SDC mode indicator, and performs depth prediction. The depth value of the current block is restored using the value and the restored depth residual.

  The SDC mode indicator according to the present invention indicates a flag indicating whether the current block is coded in the SDC mode. The SDC mode is a depth residual for a plurality of samples in the current block. It means a method of residual coding.

  When the SDC mode indicator according to the present invention indicates that the current block is coded in the SDC mode, the residual coding information is used to restore the depth residual of the current block.

  The residual coding information according to the present invention includes depth absolute absolute values and depth residual code information.

  The depth residual according to the present invention means a difference between the average value of the depth values of the current block and the average value of the depth prediction values of the current block.

  The depth residual according to the present invention means an average value of the depth residual of the i-th sample obtained from the difference between the depth value of the i-th sample of the current block and the depth predicted value of the i-th sample.

  If the SDC mode indicator according to the present invention indicates that the current block is coded in SDC mode, the depth residual is restored using the depth lookup table.

  The depth residual according to the present invention derives a difference index using the absolute value of the depth residual and the sign information of the depth residual, obtains the depth prediction average value of the current block, and obtains the depth prediction average value and the depth lookup. A prediction index is obtained using a table, a table depth value corresponding to an index obtained from the sum of the prediction index and the difference index is obtained from the depth lookup table, and the obtained table depth value and depth prediction average value are obtained. It is restored through the difference between and.

  The prediction index according to the present invention is set as a table index assigned to the table depth value that minimizes the difference between the depth prediction average value and the table depth value in the depth lookup table.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, and redundant description of the same or similar components regardless of reference numerals will be omitted. Also, the terms and words used in this specification and claims should not be construed to be limited to ordinary or lexicographic meanings, so that the inventor can best explain his invention. Based on the principle that the term concept can be appropriately defined, it should be interpreted as a meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configurations shown in the drawings are only the most preferred embodiments of the present invention and do not represent all the technical ideas of the present invention. It should be understood that there may be various equivalents and variations that can be substituted at this time.

  FIG. 1 is a schematic block diagram of a video decoder 100 according to an embodiment to which the present invention is applied.

  Referring to FIG. 1, the video decoder 100 may include a parsing unit 110, a residual restoration unit 120, an intra prediction unit 130, an in-loop filter unit 140, a decoded picture buffer unit 150, and an inter prediction unit 160.

  The parsing unit 110 can receive a bitstream including multi-view texture data. In addition, when the depth data is necessary for coding the texture data, a bit stream including the depth data can be further received. The texture data and depth data input at this time may be transmitted in one bit stream or may be transmitted in separate bit streams. In addition, when the received bit stream is multi-view related data (eg, 3-Dimensional Video), the bit stream may further include camera parameters. The camera parameters may include intrinsic camera parameters and non-intrinsic camera parameters, and the intrinsic camera parameters include focal length, aspect ratio, and main ratio. A point (principal point) or the like can be included, and the non-unique camera parameter can include information on the position of the camera in the world coordinate system.

  The parsing unit 110 performs parsing in units of NAL to decode the received bitstream, thereby coding information for predicting video images (for example, block partition information, intra prediction mode, motion information, reference index). Etc.) and coding information (for example, quantized transform coefficients, absolute values of depth residuals, depth residual code information, etc.) corresponding to video video residual data can be extracted.

  Residual restoration unit 120 obtains a scaled transform coefficient by scaling the quantized transform coefficient using the quantization parameter, and restores the residual data by inversely transforming the scaled transform coefficient. be able to. Alternatively, the residual restoration unit 120 can restore the residual data using the absolute value of the depth residual and the code information of the depth residual, which will be described later with reference to FIGS. 3 to 7. . Further, the quantization parameter for the depth block can be set in consideration of the complexity of the texture data. For example, a low quantization parameter can be set when the texture block corresponding to the depth block is a high-complexity region, and a high quantization parameter can be set when the texture block is a low-complexity region. The complexity of the texture block can be determined based on a difference value between pixels adjacent to each other in the restored texture picture as shown in Equation 1.

  In Equation 1, E indicates the complexity of the texture data, C indicates the restored texture data, and N may indicate the number of pixels in the texture data region for which the complexity is to be calculated. Referring to Equation 1, the complexity of the texture data is the difference between the texture data corresponding to the (x, y) position and the texture data corresponding to the (x-1, y) position, and the (x, y) position. Can be calculated by using a difference value between the texture data corresponding to the texture data and the texture data corresponding to the (x + 1, y) position. Further, the complexity can be calculated for each of the texture picture and the texture block, and the quantization parameter can be derived as shown in Equation 2 below using this.

  Referring to Equation 2, the quantization parameter for the depth block can be determined based on the ratio between the complexity of the texture picture and the complexity of the texture block. α and β may be variable integers derived at the decoder, or may be integers predetermined within the decoder.

  The intra prediction unit 130 can perform intra prediction using adjacent samples adjacent to the current block and the intra prediction mode. Here, the adjacent sample may be a sample located on the left side, the lower left side, the upper level, or the upper right side of the current block and the restoration is completed before the current block. Also, the intra prediction mode may be extracted from the bitstream, and may be derived based on at least one intra prediction mode of the left adjacent block or the upper adjacent block of the current block. Further, the intra prediction mode of the depth block may be derived from the intra prediction mode of the texture block corresponding to the depth block.

  The inter prediction unit 160 may perform motion compensation on the current block using the reference picture and motion information stored in the decoded picture buffer unit 150. In this specification, motion information can be understood as a broad concept including a motion vector and a reference index. Further, the inter prediction unit 160 can perform motion compensation through temporal inter prediction. Temporal inter prediction may refer to inter prediction using reference pictures located at the same viewpoint and different time zones as the current block. In addition, in the case of a multi-view video captured by a plurality of cameras, inter-view inter prediction as well as temporal inter prediction can be used. Inter-view inter prediction may mean inter prediction using a reference picture located at a different view from the current block.

  The in-loop filter unit 140 may apply an in-loop filter to each coded block in order to reduce the block distortion phenomenon. The filter can improve the picture quality of the decoded picture by smoothing the edge of the block. The filtered texture picture or depth picture can be output or stored in the decoded picture buffer unit 150 for use as a reference picture. On the other hand, since the characteristics of the texture data and the characteristics of the depth data are different from each other, when the texture data and the depth data are coded using the same in-loop filter, the coding efficiency may be lowered. Therefore, a separate in-loop filter for depth data can be defined. Hereinafter, a region-based adaptive loop filter and a trilateral loop filter will be described as in-loop filtering methods capable of efficiently coding depth data.

  In the case of a region-based adaptive loop filter, whether to apply a region-based adaptive loop filter can be determined based on the variation of depth blocks. Here, the change amount of the depth block can be defined as a difference between the maximum pixel value and the minimum pixel value in the depth block. Whether or not to apply the filter can be determined through a comparison between the change amount of the depth block and a predetermined threshold value. For example, if the amount of change in the depth block is greater than or equal to a predetermined threshold, it means that the difference between the maximum and minimum pixel values in the depth block is large, so region-based adaptation It can be determined that the dynamic loop filter is applied. Conversely, when the depth change amount is smaller than a predetermined threshold value, it can be determined that the region-based adaptive loop filter is not applied. When applying the filter according to the result of the comparison, the pixel value of the filtered depth block can be derived by applying a predetermined weight value to the neighboring pixel value. Here, the predetermined weight value may be determined based on a positional difference between the currently filtered pixel and the neighboring pixel and / or a difference value between the currently filtered pixel value and the neighboring pixel value. Also, the neighboring pixel value may mean any one of pixel values included in the depth block, excluding the pixel value that is currently filtered.

  Trilateral loop filters are similar to region-based adaptive loop filters, but differ in that they additionally consider texture data. Specifically, the trilateral loop filter can compare the following three conditions and extract the depth data of adjacent pixels that satisfy the following conditions.

  Condition 1 is to compare the positional difference between the current pixel (p) and the adjacent pixel (q) in the depth block with a predetermined parameter, and Condition 2 is the depth data of the current pixel (p) The difference between the depth data of the adjacent pixel (q) is compared with a predetermined parameter, and condition 3 is that the difference between the texture data of the current pixel (p) and the texture data of the adjacent pixel (q) It is to compare with a predetermined parameter. Neighboring pixels satisfying the three conditions can be extracted, and the current pixel (p) can be filtered by an intermediate value or an average value of these depth data.

  The decoded picture buffer unit 150 plays a role of storing or releasing previously coded texture pictures or depth pictures for inter prediction. At this time, the frame_num and the POC (Picture Order Count) of each picture can be used to store or release them in the decoded picture buffer unit 150. Further, in depth coding, there are depth pictures that have different viewpoints from the current depth picture in the previously coded pictures. In order to use such a picture as a reference picture, a viewpoint for identifying the viewpoint of the depth picture is used. Identification information can also be used. The decoded picture buffer unit 150 can manage reference pictures using an adaptive memory management method (Memory Management Method) and a moving window method (Sliding Window Method) in order to realize inter prediction more flexibly. . This is because the memory of the reference picture and the non-reference picture is unified and managed as one memory, and is efficiently managed with a small amount of memory. In depth coding, a depth picture may be marked with a separate display in order to distinguish it from a texture picture in the decoded picture buffer unit, and an identifier for identifying each depth picture may be used in the marking process.

  FIG. 2 is an internal block diagram of a broadcast receiver to which a video decoder is applied, which is an embodiment to which the present invention is applied.

  The broadcast receiver according to the present embodiment is for receiving a terrestrial broadcast signal and reproducing a video. The broadcast receiver can generate three-dimensional content using the received depth related information. The broadcast receiver includes a tuner 200, a demodulation / channel decoder 202, a transport demultiplexing unit 204, a packet demultiplexing unit 206, an audio decoder 208, a video decoder 210, a PSI / PSIP processing unit 214, a 3D rendering unit 216, a formatter 220, and A display unit 222 may be included.

  The tuner 200 selects and outputs a broadcast signal of any one channel selected by the user from among a large number of broadcast signals input via an antenna (not shown).

  The demodulation / channel decoder 202 demodulates the broadcast signal from the tuner 200, performs error correction decoding on the demodulated signal, and outputs a transport stream TS.

  The transport demultiplexing unit 204 demultiplexes the transport stream, separates the video PES and the audio PES, and extracts PSI / PSIP information.

  The packet canceling unit 206 cancels the packet for the video PES and the audio PES to restore the video ES and the audio ES.

  The audio decoder 208 decodes the audio ES and outputs an audio bit stream. The audio bit stream is converted into an analog audio signal by a digital-analog converter (not shown), amplified by an amplifier (not shown), and then output through a speaker (not shown).

  The video decoder 210 decodes the video ES to restore the original video, which has been described with reference to FIG. The decoding process of the audio decoder 208 and the video decoder 210 may be performed based on a packet ID (PID) confirmed by the PSI / PSIP processing unit 214. In the decoding process, the video decoder 210 may extract depth information. Further, additional information necessary for generating a virtual camera viewpoint video, for example, camera information, or information for estimating a region (Occlusion) obstructed by a relatively front object (for example, an outline of an object, etc.) , Geometric information, object transparency information and color information) can be extracted and provided to the 3D rendering unit 216. However, in another embodiment of the present invention, the depth information and / or the additional information may be separated by the transport demultiplexing unit 204.

  The PSI / PSIP processing unit 214 receives the PSI / PSIP information from the transport demultiplexing unit 204, parses it, and stores it in a memory (not shown) or a register to base the stored information. As a broadcast is played.

  The 3D rendering unit 216 can generate color information, depth information, and the like at the virtual camera position using the restored video, depth information, additional information, and camera parameters. In addition, the 3D rendering unit 216 generates a virtual video at the virtual camera position by performing 3D warping using the restored video and the depth information on the restored video. In the present embodiment, the 3D rendering unit 216 is described as being configured as a separate block from the video decoder 210. However, this is only an example, and the 3D rendering unit 216 is included in the video decoder 210. It may be broken.

  The formatter 220 formats the video restored by the decoding process, that is, the video shot by the actual camera and the virtual video generated by the 3D rendering unit 216 according to the display method in the receiver, A 3D image is displayed via the display 222. Here, the depth information and the virtual video at the virtual camera position by the 3D rendering unit 216 and the video formatting by the formatter 220 may be selectively performed in response to a user command. That is, the viewer may operate the remote controller (not shown) so that the synthesized video is not displayed, and can specify the time point when the video synthesis is performed.

  As described above, the depth information is used in the 3D rendering unit 216 to generate the 3D video, but may be used in the video decoder 210 as another embodiment.

  FIG. 3 is a flowchart illustrating a process of restoring the depth value of the current block according to an embodiment to which the present invention is applied.

  Referring to FIG. 3, the depth prediction value of the current block can be obtained (S300). Specifically, when the current block is coded in the intra mode, the depth prediction value of the current block can be obtained using adjacent samples adjacent to the current block and the intra prediction mode of the current block. Here, the intra prediction mode may include a Planar mode, a DC mode, and an Angular mode. Alternatively, when the current block is coded in the inter mode, the depth prediction value of the current block can be obtained using the motion information of the current block and the reference picture.

  The depth residual can be restored for each sample of the current block according to the SDC mode indicator (S310). Here, the SDC mode indicator may mean a flag indicating whether the current block is coded in the SDC mode. The SDC mode may mean a method of coding depth residuals for a plurality of samples in the current block into one residual. Also, the depth residual can be restored only when the current block is not coded in the skip mode. This is because there is no residual data in the skip mode.

  Specifically, if the current block is not coded in the SDC mode according to the SDC mode indicator, a quantized transform coefficient can be obtained from the bitstream. The acquired quantized transform coefficient can be scaled using the quantization parameter, and the scaled transform coefficient can be inversely transformed to restore the depth residual.

  On the other hand, if the current block is coded in the SDC mode according to the SDC mode indicator, the depth residual of the current block can be restored using the residual coding information. Here, the residual coding information may include a depth residual absolute value and a depth residual code information. On the other hand, the residual coding information may be described separately when coded without using a depth look-up table (DLT) and when coded with a depth look-up table. . The depth lookup table is for improving coding efficiency by coding an index by assigning an index corresponding to the depth value without coding the depth value as it is. Therefore, the depth lookup table may be a table that defines table depth values and a table index corresponding to each table depth value. The table depth value may include at least one or more depth values that cover the range of the depth minimum and depth maximum of the current block. Further, the table depth value may be encoded by an encoder and transmitted via a bit stream, or a predetermined value in a decoder may be used.

  Hereinafter, a method for encoding residual coding information and a method for restoring depth residual using residual coding information when the depth look-up table is not used will be described with reference to FIGS. 4 and 5, respectively. . Also, a method of encoding residual coding information and a method of restoring depth residual using residual coding information when using a depth lookup table will be described with reference to FIGS. 6 and 7, respectively.

  The depth value of the current block can be restored using the depth prediction value obtained in step S300 and the depth residual restored in step S310 (S320). For example, the depth value of the current block may be derived from the sum of the depth prediction value and the depth residual. Also, the depth value of the current block can be derived for each sample.

  FIG. 4 shows an embodiment to which the present invention is applied, and shows a method of encoding residual coding information when a depth look-up table is not used.

  1. First method

  A first method according to the present invention is a method of obtaining a depth residual of a current block through a difference operation after an average calculation of a depth value (original depth value) of the current block and a depth prediction value of the current block (difference prediction value). It is.

  Referring to FIG. 4A, the average value (DCorig) of the depth values of the current block is obtained. Then, an average value (DCpred) of depth prediction values of the current block is acquired. The depth residual (DCres) can be acquired through the difference between the average value of the acquired depth values and the average value of the depth prediction values. The depth residual can be coded into the absolute value (DCabs) of the depth residual and the sign information (DCsign) of the depth residual and transmitted to the decoder.

  2. Second method

  The second method according to the present invention is a method of obtaining the depth residual of the current block through an average operation after calculating the difference between the depth value of the current block and the depth prediction value.

  Referring to FIG. 4B, through the difference calculation between the depth value (Origi) of the i-th sample of the current block and the corresponding depth prediction value (Predi) of the i-th sample, Depth Residual (Resi) can be earned. Here, if the current block is an NxN block, i is greater than or equal to 0 and less than or equal to N2-1, which may locate the sample. Then, the depth residual (DCres) of the current block can be obtained through an average operation between N2 depth residuals. Similarly, the depth residual can be coded into the absolute value (DCabs) of the depth residual and the sign information (DCsign) of the depth residual and transmitted to the decoder.

  As described above, an average operation can be used to code the depth residual of the current block into one depth residual according to the SDC mode. However, the present invention is not limited to this, and it is a matter of course that one depth residual can be obtained from the maximum value, the minimum value, or the mode value among the plurality of depth residuals of the current block.

  FIG. 5 shows an embodiment to which the present invention is applied, and illustrates a method for obtaining a depth residual of a current block using residual coding information when a depth lookup table is not used. .

  Depth-residual absolute value and depth-residual code information can be extracted from the bitstream (S500).

  The depth residual of the current block can be derived using the absolute value of the depth residual extracted in step S500 and the code information of the depth residual (S510). Here, when the absolute value and the sign information of the depth residual are coded by the first method described with reference to FIG. 4, the depth residual is an average of the depth value of the current block and the depth prediction value of the current block. It can be defined as the difference from the average value. On the other hand, when the absolute value and sign information of the depth residual are coded according to the second method described with reference to FIG. 4, the depth residual is the depth value of the i-th sample and the depth of the i-th sample of the current block. It can be defined as the average value of the depth residual of the i-th sample obtained from the difference from the predicted value.

  FIG. 6 shows an embodiment to which the present invention is applied and shows a method of encoding residual coding information when a depth look-up table is used.

  Referring to FIG. 6, the depth average value (DCorig) of the current block can be obtained. Here, the depth average value may mean an average value of the depth values of a plurality of samples included in the current block.

  Then, a depth index (Iorig) can be obtained using the obtained depth average value (DCorig) and the depth lookup table of the current block.

  Specifically, the table depth value in the depth look-up table corresponding to the depth average value (DCorig) can be determined. The determined table depth value may mean a table depth value that minimizes a difference between the depth average value (DCorig) and the table depth value in the depth lookup table. Then, the table index assigned to the determined table depth value can be set as the depth index (Iorig).

  Meanwhile, the depth prediction value of the current block can be obtained. The depth prediction value can be acquired in any one of the intra mode and the inter mode. An average value between depth prediction values of a plurality of samples included in the current block (hereinafter referred to as depth prediction average value (DCpred)). Can be earned.

  The prediction index (Ipred) can be obtained using the depth prediction average value (DCpred) and the depth lookup table of the current block. Specifically, the table depth value in the depth lookup table corresponding to the depth prediction average value (DCpred) can be determined. The determined table depth value may mean a table depth value that minimizes a difference between a depth prediction average value (DCpred) and a table depth value in the depth lookup table. Then, the table index assigned to the determined table depth value can be set as the prediction index (Ipred).

  Thereafter, a difference index (Ires) between the acquired depth index (Iorig) and the prediction index (Ipred) can be acquired. The difference index (Ires) may be encoded into residual coding information including depth absolute values (DCabs) and depth residual code information (DCsign), as in the case of not using a depth lookup table. However, the absolute value of the depth residual can mean the absolute value of the difference index (Ires), and the code information of the depth residual can mean the sign of the difference index (Ires). In other words, if no depth lookup table is used, the depth residual is coded to the value of the sample domain, while if using the depth lookup table, the depth residual can be coded to the value of the index domain.

  FIG. 7 shows an embodiment to which the present invention is applied and shows a method for restoring depth residual using residual coding information when using a depth look-up table.

  Residual coding information can be obtained from the bitstream. Residual coding information may include depth absolute absolute values (DCabs) and depth residual code information (DCsign). The difference index (Ires) can be derived using the absolute value (DCabs) of the depth residual and the sign information (DCsign) of the depth residual.

  Meanwhile, coding information (for example, intra prediction mode, motion information, etc.) for predicting the current block can be further acquired from the bitstream. Depth prediction values can be obtained for each sample of the current block using coding information, and an average value of the obtained depth prediction values, that is, a depth prediction average value (DCpred) can be obtained.

  The prediction index (Ipred) can be obtained using the depth prediction average value (DCpred) and the depth lookup table of the current block. Here, the prediction index (Ipred) is a table assigned to a table depth value that minimizes the difference between the depth prediction average value (DCpred) and the table depth value in the depth lookup table, as described with reference to FIG. Can be set as an index.

  Thereafter, the depth residual can be restored using the prediction index (Ipred), the difference index (Ires), and the depth lookup table.

  For example, a table depth value (Idx2DepthValue (Ipred + Ires)) corresponding to the index obtained from the sum of the prediction index (Ipred) and the difference index (Ires) can be obtained from the depth lookup table. Thereafter, the depth residual of the current block can be restored from the difference between the acquired table depth value and the depth predicted average value (DCpred).

  In each of the embodiments described above, the components and features of the present invention are combined in a predetermined form. Each component or feature must be considered optional unless stated otherwise. Each component or feature may be implemented in a form that is not combined with other components or features. In addition, it is possible to configure an embodiment of the present invention by combining some components and / or features. The order of operations described in the embodiments of the present invention can be changed. Some configurations or features of one embodiment can be included in other embodiments, or can be replaced with corresponding configurations or features of other embodiments.

  The present invention can be used to encode or decode a video signal.

Claims (8)

Obtaining a depth prediction for the current block;
Restoring depth residual by sample of the current block according to an SDC mode indicator;
Restoring the depth value of the current block using the depth prediction value and the restored depth residual.
The SDC mode indicator is a flag indicating whether or not the current block is coded in the SDC mode. The SDC mode is a depth residual for a plurality of samples in the current block. A video signal processing method characterized in that it means a method of residual coding. If the SDC mode indicator indicates that the current block is coded in SDC mode, restoring the depth residual is:
Extracting residual coding information from the bitstream;
Using the extracted residual coding information to derive the depth residual of the current block;
The video signal processing method according to claim 1, wherein the residual coding information includes an absolute value of depth residual and code information of depth residual.   The method of claim 2, wherein the induced depth residual means a difference between an average value of depth values of the current block and an average value of depth prediction values of the current block.   The derived depth residual means an average value of the depth residual of the i-th sample obtained from the difference between the depth value of the i-th sample of the current block and the depth predicted value of the i-th sample. The video signal processing method according to claim 2.   The method of claim 1, wherein the depth residual is restored using a depth lookup table if the SDC mode indicator indicates that the current block is coded in SDC mode. . The step of restoring the depth residual is:
Obtaining residual coding information from the bitstream, wherein the residual coding information includes an absolute value of depth residual and code information of depth residual;
Deriving a difference index using the absolute value of the depth residual and the sign information of the depth residual; and
Obtaining a depth prediction average value of the current block, wherein the depth prediction average value means an average value of the acquired depth prediction values;
Obtaining a prediction index using the depth prediction average value and the depth lookup table;
Obtaining a table depth value corresponding to an index obtained from the sum of the prediction index and the difference index from the depth lookup table;
The video signal processing method according to claim 5, further comprising: restoring a depth residual of the current block from a difference between the acquired table depth value and the depth prediction average value.   The video according to claim 6, wherein the prediction index is set as a table index assigned to a table depth value that minimizes a difference between the depth prediction average value and a table depth value in the depth lookup table. Signal processing method. An inter prediction unit that obtains the depth prediction value of the current block;
A residual restoring unit that restores depth residual for each sample of the current block according to an SDC mode indicator;
A depth restoration unit that restores the depth value of the current block using the depth prediction value and the restored depth residual, and
The SDC mode indicator is a flag indicating whether or not the current block is coded in the SDC mode. The SDC mode is a depth residual for a plurality of samples in the current block. A video signal processing apparatus, characterized in that it means a method of residual coding.

Images